DATA
BRIEF
16-06
What We’re Learning: Work-Study Program
A Data Update from the Wisconsin HOPE Lab
October 19, 2016
For over fifty years the Federal Work-Study Program has funded part-time employment to help
college students finance their education.1 The program is intended to assist low-income students
in particular. However, recent evidence indicates that only a fraction—less than 20 percent—
of the lowest-income students obtain Federal Work-Study funded employment.2 As a result,
advocates and policymakers are questioning whether the program is fulfilling its mission to serve
low-income college students. Another of the program goals is that Work-Study employment
would serve the public interest and be relevant to the student’s academic goals.3 Student
advocates also question whether these are being met.4
This data brief describes empirical evidence on college students who are employed in Federal
Work-Study jobs, using survey data from an ongoing study of financial aid recipients in
Wisconsin.5 Rather than focusing on how much Federal Work-Study employment is tied to actual
financial need—a topic better addressed using national data —we use student survey responses
to characterize Federal Work-Study employment more fully than is currently possible with
Campus Compact. (2015). “A Brief History of the Federal Work-Study Program.” Available at http://compact.org.
O’Sullivan, R. & Setzer, R. (2014). “A Federal Work-Study Reform Agenda to Better Serve Low-Income
Students.” Young Invincibles.
1
2
Federal Studen Aid website: https://studentaid.ed.gov/sa/types/work-study
O’Sullivan, R. & Setzer, R. (2014). “A Federal Work-Study Reform Agenda to Better Serve Low-Income
Students.” Young Invincibles.
3
4
More information on the study can be found at the end of this brief, and on our website at http://www.wihopelab.
com/projects/Impact-STEM-success.html
5
WISCONSIN
HOPE
LAB
|
Date
Brief
16-06
1
measures from national surveys. We also compare Federal Work-Study jobs to other types of
employment held by students in our sample in terms of schedules, earnings, and job relevance to
academic and career goals.
The Federal Work-Study Program
The Federal Work-Study (Work-Study) program began in 1964 as part of the Economic
Opportunity Act. The goal was to help alleviate the burden of college costs for low-income
students by providing subsidized part-time work. Jobs funded by Work-Study are meant to be
as closely related to the students’ educational goals as possible, and can be located either oncampus or off-campus with a non-profit agency, government agency, or for-profit organization.
Work-Study is a campus-based financial aid program, meaning that financial aid offices are
given broad discretion to make funding allocation decisions. Typically, offices offer Work-Study
funds on a first-come, first-served basis to help cover unmet need left in students’ financial
aid packages after grants and scholarships. Students are then required to find a job with a
participating employer where they earn wages subsidized by their Work-Study award.
For the 2011-2012 school year approximately 6% of students were awarded Federal Work-Study,
with average Work-Study funding per student of $2,300 annually.6 In our study, the rate of WorkStudy employment was somewhat higher, but holding a Work-Study position was still uncommon.
Just 11% of sample students reported working at a Work-Study job during their second year in
college, while nearly six times that number (60%) reported working at a job unaffiliated with the
program. Among students eligible to receive a Federal Pell Grant, 12% held a Work-Study job,
compared to 8% of non-Pell-eligible students.
What is a Typical Weekly Schedule at a Federal Work-Study
Job?
On average, students in our study employed in Work-Study jobs alone worked 11 hours per
week. By comparison, those at working only at non-Work-Study jobs reported working double
that number at 22 hours per week (Figure 1). Work-Study awards are typically modest and
students are usually not allowed to earn in excess of the amount of Work-Study awarded.7 Thus
the large gap in hours worked between Work-Study and non-Work-Study jobs. This limitation
on Work-Study hours may also help to explain why 22% of students with Work-Study jobs in
our sample also worked a second job. This group of students who combined Work-Study and
National Center for Education Statistics. (2013). “2011-12 National Postsecondary Student Aid Study.” Table 2.
Available at http://nces.ed.gov/pubs2013/2013165.pdf.
7
At the average Work Study award amount of $2,300 per year, and assuming a 75% to 25% cost share with
the employer, a student earning $8.00 per hour would be limited to working around 12-13 hours per week during
the academic year (based on two 15-week semesters). Based on data from “National Center for Education
Statistics. (2013).“ 2011-12 National Postsecondary Student Aid Study.” Table 2. Available at http://nces.ed.gov/
pubs2013/2013165.pdf
6
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
2
non-Work-Study employment worked a total of 26 hours per week on average—9 hours a week
at Work-Study jobs and 17 hours a week at non-Work-Study jobs. Total work hours for this
Combined group were nearly two and a half times more than for those with Work-Study jobs only
and 4 hours more than students who held only a non-Work-Study job. Such long working hours
are cause for concern, because research shows that working more than 20 hours per week is
associated with poor academic outcomes.8
We report on three groups of students: 1) Students who worked at Work-Study jobs only (n=68);
2) Students who worked at non-Work-Study jobs only (n=471); and 3) A Combined group of
students that worked both Federal Work-Study and non-Federal Work-Study jobs (n=26).
Figure 1: Hours worked per week by employment type
30
*
HOURS/WEEK
25
*
20
17
15
22
10
11
5
0
9
Work-Study Only
Non-Work-Study Only
Work-Study
Combined Group
Non-Work-Study
* Significantly different from Work-Study Only group at 0.05 level
A potential advantage of Work-Study jobs is the ability to work a schedule more conducive to
course attendance and studying. Looking at the times of day students worked, students working
only Work-Study jobs were seven percentage points more likely than those with non-Work-Study
jobs only to work daytime hours between 8 a.m. and 6 p.m. (Figure 2). Daytime schedules are
advantageous in that they free up evenings for study and rest. They could, however, interfere
with course schedules. Students working only at Work-Study jobs were less likely to work
evenings (6 p.m. to 10 p.m.) compared to those with non-Work-Study positions only. However,
Bozick, R. (2007). Making it through the first year of college: The role of students’ economic resources,
employment, and living arrangements. Sociology of Education, 80(3), 261–285.
8
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
3
more than one-third (36%) of the Work-Study Only group reported working the third shift (10
p.m. to 8 a.m.), as did 39% of the non-Work-Study Only group. Working into the wee hours of
the morning likely cut into students’ sleep schedules, making it difficult for them to engage fully in
their classes. All students in the Combined group worked between 8 a.m. and 6 p.m. (Figure 2).
And they were more likely than students in the other groups to work evenings (87%) or during the
third shift (53%). These students worked the most hours per week on average and it appears that
to balance work and school they worked later into the night.
Figure 2: Time of day worked by employment type
95%
Day (8 a.m. to 6 p.m.)
88%*
48%
Night (6 p.m. to 10 p.m.)
36%
39%
Thrid Shift ( 10 p.m. to 8 a.m.)
71%*
100%*
87%*
53%*
0% 10% 20% 3% 40% 50% 60% 70% 80% 90% 100%
Work-Study
Non-Work-Study
Combined Group
Note: * Significantly different from Work-Study Only at the 0.05 level.
Beyond the time of day worked, students reported that Work-Study jobs were more compatible
with college life in terms of schedule predictability, compatibility of work and course schedules,
and the ability to study on the job.9
Although the vast majority of students reported working daytime jobs, this does not appear to
have interfered with their course schedules. Eighty-nine percent of Work-Study Only students
reported that work rarely interfered with their course schedule (Figure 3). This rate was
somewhat lower for the Non-Work-Study Only group, 77% of whom reported lack of conflict
Original survey question asked students to separately rate their FWS and/or non-FWS job in terms of how
much benefit they derived from the job in each area (e.g. “How much are you able to do school work while at
this job?”). Original answers were on a five-point scale where: 1=Not at all; 2=A little; 3=Somewhat; 4=Quite
a bit; and 5=A great deal. We report percentages of students whose response was either of the top two scale
categories (“Quite a bit” or “A great deal”).
9
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
4
between work and course schedules. The Combined group reported high rates of schedule
compatibility between work and class for both types of employment (Table 1).
Students reported more consistent work hours for their Work-Study jobs. Students in the WorkStudy Only group were 26 percentage points more likely to report having a consistent work
schedule from week to week compared to their peers with non-Work-Study positions (Figure 3).
And students in the Combined group were much more likely to report consistent work schedules
for their Work-Study jobs compared to their non-Work-Study jobs (Table 1). Such schedule
predictability may allow students to more easily plan courses and study times around their work.
In contrast, work schedules that change from week to week may limit students’ ability to plan
ahead, thus inhibiting their ability to perform academically.
Another schedule-related benefit of Work-Study positions was the ability to study while on the
job. Thirty-eight percent of students in the Work-Study Only group reported being able to do
school work on the job compared to just 23% who said the same from the Non-Work- Study Only
group (Figure 3). Thirty-eight percent of the Combined group reported being able to study at their
Work-Study job, but only 9% of this group reported being able to do the same at their non-WorkStudy job (Table 1).
Figure 3: Work schedule compatibility with academics
38%
Able to do school work while at job
23%*
89%
Rarely interferes with course schedule
77%*
73%
Work schedule conistent
47%*
0% 10% 20% 3% 40% 50% 60% 70% 80% 90% 100%
Work-Study Only
Non-Work-Study Only
Note: * Significantly different from Work-Study Only at the 0.05 level.
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
5
Table 1. Work schedule compatibility with academics:
Students working both Work-Study and non-Work-Study jobs
Work-Study
Job
Non-WorkStudy Job
Both Jobs
Total
Able to do school
work while at job
35%
6%
3%
44%
Rarely interferes
with course schedule
11%
13%
66%
90%
Work schedule
consistent
43%
3%
27%
73%
How much do students earn at their Federal Work-Study job?
On average, students in the Work-Study Only group earned $8.47 per hour (Figure 3). This was
about two dollars less than the average wage for the Non-Work-Study Only group ($10.56 per
hour). Considering that students with only an Work-Study job worked an average of 11 hours a
week while students with a non-Work-Study job worked an average of 22 hours per week; and in
light of the wage difference between these two groups; students working Work-Study jobs earned
approximately $4,175 less than those working non-Work-Study jobs over the course of the
school year.10 Given this difference, it would be unsurprising if students offered Work-Study funds
in their aid package decided to forgo a Work-Study position in order to find a job with higher pay
and longer hours elsewhere.
Students in the Combined group earned lower pay at both jobs compared to students working
either type of job alone. They earned an average of $7.91 at their Work-Study jobs and $8.82
at their non-Work-Study jobs. These lower pay rates may help explain why students in the
Combined group sought additional work. Combined group students earned roughly $2,100 at
their Work-Study positions and boosted their earnings by approximately $4,500 for the school
year by also working non-Work-Study jobs.
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
6
Figure 3: Hourly wage by employment type
$12.00
*
DOLLARS/HOUR
$10.00
*
$8.82
$8.00
$6.00
$10.56
$8.47
$4.00
$7.91
$2.00
$0.00
Work-Study Only
Non-Work-Study Only
Work-Study
Combined Group
Non-Work-Study
Note: * Significant different from Work-Study Only at the 0.05 level
Are Work-Study jobs academically or career relevant?
Finally, we wanted to know if Work-Study was delivering on the promise of work that was
academically relevant. Our analyses indicate that Work-Study jobs offered no benefits or
even negative benefits compared to non-Work-Study positions along this dimension. Students
employed in Work-Study jobs only were 10 percentage points less likely to report that their job
was related to their academic major than those with non-Work-Study jobs only.11 Only 5% of the
Work-Study Only group reported career relevant employment, compared to 23% of those in the
Non-Work-Study Only group. Finally, only about one-third of students working only Work-Study
jobs reported that they learned new skills on the job, compared to half of students in the NonWork-Study Only group.
Calculation based on 30-week academic year.
Original survey question asked students to separately rate their FWS and/or non-FWS job in terms of how
much benefit they derived from the job in each area (e.g. “How much does your job relate to the work you
would like to do for a career?”). Original answers were on a five-point scale where: 1=Not at all; 2=A little;
3=Somewhat; 4=Quite a bit; and 5=A great deal. We report percentages of students whose response was either
of the top two scale categories (“Quite a bit” or “A great deal”).
10
11
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
7
Figure 4: Academic and career relevance by employment type
13%
Related to academic major
23%*
5%
Relevant to career goals
23%*
32%
Learned new skills
50%*
0%
20%
Work-Study Only
40%
60%
Non-Work-Study Only
Note: * Significant from Work-Study Only at the 0.05 level
Students in the Combined group reported more even rates of academic and career relevance
across job types (Figure 5). The Combined group more often reported learning new skills at their
Work-Study jobs than at their non-Work-Study jobs; a trend opposite that observed above for the
groups working only one job type. These results provide additional clues as to why students in
the Combined group chose to keep their Work-Study job and take on a second job. They appear
to have gotten good experience from their Work-Study jobs, but likely took another job to bolster
their income.
Table 2. Academic and career relevance by employment type:
Students working both FWS and non-FWS jobs
Work-Study
Job
Non-WorkStudy Job
Both Jobs
Total
Relevant to
academic goals
14%
12%
5%
32%
Relevant to career
goals
14%
12%
5%
32%
Learned new
skills
24%
17%
28%
68%
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
8
Conclusion
The data presented in this brief suggest that the Federal Work-Study Program provided students
in our sample with employment that better accommodated their college course schedules and
provided predictable work schedules. Work-Study jobs also afforded the opportunity to do school
work on the job. This finding is broadly consistent with recent research indicating that WorkStudy jobs are advantageous to some students by providing benefits such as flexible schedules
and short work commutes, which can promote academic success by freeing up time for studying
and course attendance.12
However, we also found that students in Work-Study jobs were likely to earn lower wages relative
to working outside the program—about 20% less on average. This is perhaps why around one in
five students in our study who had a Work-Study job also took on additional employment. Further,
students who took Work-Study jobs were no more likely than those in non-Work-Study positions
to feel that they were gaining new work skills or experience relevant to their academic and career
goals. In fact, Work-Study-employed students were generally less likely to report these types of
benefits from their job. If nothing else, such evidence should give pause to the notion that the
Federal Work-Study Program adds value over non-program employment by focusing on the
career readiness of program participants. Reforms to the program should focus on these issues,
while also addressing issues of overall funding and targeting recommended by others.
Scott-Clayton, J. & Minaya, V. (2016). Should student employment be subsidized? Conditional counterfactuals
and the outcomes of work-study participation. Economics of Education Review, 52, 1-18.
12
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
9
Data discussed in this brief were collected as part of a larger research project
investigating the impact of financial aid on persistence in science, technology,
engineering, and math (STEM) majors. Students were recruited in the fall of 2014
at seven campuses of the University of Wisconsin system, two public technical
colleges, and one private non-profit technical college. Participants were mostly
first-time entering students, and in order to be eligible they had to be Wisconsin
residents, be enrolled in at least one credit, have an EFC of $10,314 or less (200%
of the Pell cut-off for the 2014-15 academic year), have at least $1000 of unmet
need, have demonstrated a modest interest in STEM fields, and have test scores
indicating they would not require remediation in math.
Analyses in this report are based on questions from a comprehensive survey
sent to all study eligible students in the fall of 2015 when most were entering their
second year of college. The analysis sample consists of 842 students with an
average age of 19 at the time of college entry. Fifty percent are female and 21%
are racial or ethnic minorities. About two-thirds (63%) of the sample was Pelleligible at the time of the survey, and 23% had an expected family contribution of
zero indicating a high level of economic disadvantage.
WISCONSIN
HOPE
LAB
|
Data
Brief
16-06
10